From the AGA Journals

Network meta-analysis ranks first-line H. pylori regimens

View on the News

Weighing options is important

In this perspective, the network meta-analysis by Rokkas and colleagues is very important: The purpose of this study is not only to identify those regimens with the highest treatment success in comparison but also stratifies for world regions and time-shift aspects. The key value of the network approach, however, is the ability for indirect comparisons, as presented here. Using the surface under the cumulative ranking values, vonoprazan-based triple therapy may be the most promising candidate for the future, non–bismuth quadruple and R-hybrid therapies are also suitable.

Gerhard G. Treiber, MD, AGAF, is with department of internal medicine at Saarland University Hospital, Homburg, Germany.

Dr. Gerhard G. Treiber

So what is the take-home message from this paper? Unfortunately, the authors could not include data concerning drug dosage and resistance. I think that emphasizing the need for antibiotic stewardship on one hand and – at the same time – telling us to still rely on local resistance knowledge (whatever this means) is not enough in 2021. Our unit routinely monitors Helicobacter pylori resistance with a polymerase chain reaction technique in each single patient, revealing rates for resistance to macrolides and fluoroquinolones of around 20%. (Cost-effectiveness advice: Take only those biopsy specimens that have turned to be positive in the rapid urease test and send them in for polymerase chain reaction testing within 72 hours; 90% success.)

In this perspective, with currently sparse vonoprazan data limited to Japan, I still prefer to go primarily for the non–bismuth quadruple therapy (56 pills to be taken in 1 week), and from my own published data, this regimen will still work if only taken for 5 days. Vice versa, in the presence of macrolide resistance, amoxicillin allergy, previous treatment failures, I go for the bismuth quadruple therapy – if I can expect good treatment compliance because proton pump inhibitor plus potassium, metronidazole, and tetracycline for 10 days can mean 140 pills. Gerhard G. Treiber, MD, AGAF, is with the department of internal medicine at Saarland University Hospital, Homburg, Germany. He has no conflicts of interest.



network meta-analysis of current first-line dual, triple, and quadruple therapies for Helicobacter pylori infection found that vonoprazan triple therapy was most effective, while standard triple therapy of a proton pump inhibitor (PPI), amoxicillin, and clarithromycin was least effective. Levofloxacin-containing triple therapy performed best in Western countries and West Asia, while reverse hybrid therapy was most effective in East Asia.

Theodore Rokkas, PhD, MD, European University of Cyprus in Engomi

Dr. Theodore Rokkas

The results “[suggest that] a new approach concerning H. pylori treatment is now needed and that the time for transitioning from trial and error to antimicrobial stewardship [of H. pylori infection] has arrived,” wrote Theodore Rokkas, PhD, MD, of the European University of Cyprus in Engomi, and colleagues. Their study was published online April 8 in Gastroenterology.

H. pylori infection is the primary cause of gastritis, peptic ulcer disease, gastric mucosa–associated lymphoid tissue lymphoma, and gastric cancer.

Since H. pylori infection was first recognized, physicians have employed a range of drugs in double, triple, and quadruple combinations to combat it.

Despite those efforts, treatment success is lower than with many other infectious diseases. A newcomer is the potassium-competing acid blocker vonoprazan, which increases efficacy of amoxicillin combination therapies and has, thereby, generated renewed interest in all combination therapies, according to the study authors. Vonoprazan is currently available in some Asian countries, but not the United States or Europe.

Current guidelines for H. pylori treatment relied on randomized controlled trials and relevant pair-wise meta-analyses, but no previous pairwise analysis has included all currently available medications, the authors noted. Network meta-analyses can help fill this evidence gap: They incorporate both direct and indirect evidence from a collection of randomized controlled trials to estimate the comparative effectiveness of three or more regimens.

The researchers conducted a network meta-analysis that included 68 randomized, controlled trials totaling 22,975 patients. The following regimens were included in the analysis: Concomitant quadruple bismuth treatment (bismuth quadruple therapy), concomitant quadruple nonbismuth treatment (nonbismuth quadruple therapy), high-dose amoxicillin double treatment (Amox-dual therapy), levofloxacin-containing treatment (Levo-therapy), reverse hybrid therapy (R-hybrid therapy), sequential quadruple treatment (sequential therapy), standard triple treatment (triple therapy), and vonoprazan-containing therapy (Vono-triple therapy).

Statistically significant results were found with Vono-triple therapy versus triple therapy (odds ratio, 3.80; 95% confidence interval, 1.62-8.94), sequential therapy versus triple therapy (OR, 1.79; 95% CI, 1.26-2.53), nonbismuth quadruple therapy versus triple therapy (OR, 2.08; 95% CI, 1.45-2.98), bismuth quadruple therapy versus triple therapy (OR, 1.47; 95% CI, 1.02-2.11), and Levo-therapy versus triple therapy (OR, 1.79; 95% CI, 1.26-2.53).

In the overall data, mean cure rates greater than 90% were seen only in Vono-triple therapy (91.4%; 95% CI, 88.5-93.5%) and R-hybrid therapy (93.6%; 95% CI, 90.4-96.8%). Cure rates were lower for Nonbismuth quadruple therapy (84.3%; 95% CI, 82.7-85.8%), Levo-therapy (83.8%; 95% CI, 82.1-85.4%), Sequential therapy (83.7%; 95% CI, 82.7-84.7%), bismuth quadruple therapy (81.3%; 95% CI, 79.5-83.1%), Amox-dual therapy (80.2%; 75.3%-84.4%), and triple therapy (75.7%; 95% CI, 74.9-76.4%). Levo-therapy performed best in Western countries (88.5%; 95% CI, 86.5-90.5%) and West Asia (88.4%; 95% CI, 84.6-91.1%). R-hybrid therapy performed best in East Asia (93.6%; 95% CI, 90.4-96.8%).

A surface under the cumulative ranking (SUCRA) value, which represents the efficacy of the intervention compared to an ideal intervention, was 92.4% for Vono-triple therapy. The second highest SUCRA value was for 68.8% for nonbismuth quadruple therapy. The SUCRA value of standard triple therapy was 4.7%.

A key limitation to the study is that Vono-triple therapy was tested only in Japan, and requires additional study in other geographic regions.

The study received support from the Department of Veteran Affairs. The authors have consulted for and received research funding from various pharmaceutical companies.

Next Article: